askerlee

I see. I didn't consider this situation as I mainly use my own embeddings. Yeah it's better to be fixed. Thanks for finding out. On Thu, May 11, 2017 at...

Just copy the test.prototxt from the ResNet101_BN_SCALE_Merged folder. They are the same (hard example mining only happens in training, so the test model is the same).

change `cls_prob[i,label]` to `cls_prob[i,int(label)]` in lib/roi_data_layer/layer.py:242.

Same here. A temporary fix is add file association: "*.mdx": "markdown" in the settings. So far it works well :)

In my case the messages repeat on the ssh console, and my workaround is to disconnect ssh and then reconnect.

Thanks for sharing. I'm surprised there's no documentation showing this usage.

Thanks for the tips. In my case, setting up a fast DNS server speeds up downloading significantly.

I met the same problem as @auroua . His fix works for me. The image used is: https://www.dropbox.com/s/esdu4vevnnk0yev/breadmix.png?dl=0

Thanks @ethanabrooks for proposing this feature. It could be used to share experimental details in academic paper submissions. Thanks Wandb staff for taking this proposal into consideration.

Hi @OvO1111 yes of course you can train without drishti and rim. I guess maybe when you modified ds_names, some kind of bug was introduced? Could you try an alternative...